The perception of risk to one's health can be closely associated to news coverage of instances of those risks. Those perceptions, in turn, can be linked to increases or decreases in risky behaviors on the part of individuals. Therefore, an important objective of cancer control efforts is to raise awareness of the risky behaviors, and make them salient enough to potentially influence individual behavior and community policy decision-making. This project seeks to measure the association between the news media's coverage of tobacco and cancer-related risks and issues and public perceptions of the need for cancer control policy and personal behavior changes. Data on media coverage will be obtained by coding cancer-related content from the U.S. news media (local newspaper, local and national television news, national news magazines) in 2002-03, already identified and collected by an ongoing study of alcohol and illicit drug use. Data on cancer related attitudes and perceptions and news media use of average citizens will be provided from the NCI's Health Information National Trend Study (HINTS), a large national probability survey of U.S. adults collected in 2003. The specific aims are: (1) identify the relative amount and type of news coverage given to cancer and cancer-related risks; (2) characterize the manner in which cancer is articulated in news coverage; (3) test for evidence that estimated risk likelihood, prevention behaviors, screening knowledge and other cancer-related behaviors are a function of news coverage; (4) test for evidence that public perceptions and attitudes are related to levels of media coverage; and (5) identify cancer-related risks and cancer control policies that are under-reported relative to their potential for leading to policy efforts aimed at reducing cancer in the United States. The design and analysis methods parallels "exposure testing" in epidemiologic studies, i.e., linking risk perceptions to exposures to various levels of news coverage that has certain content characteristics.